Versions Compared

Version Old Version 19 New Version Current
Changes made by

Gardner, I.V. (Bella)

Gardner, I.V. (Bella)

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Table of Contents
stylenone

Getting Started

giphy (4).webpImage Removedclass.gifImage Added

Natural Language Understanding (NLU) is a core component of conversational AI systems, enabling machines to interpret and act on user input in natural language. This Intent and Slot Classifier project is designed to help students understand the pipeline involved in building an NLU model that performs intent classification and slot filling. These tasks allow AI models to classify a user's goal (intent) and extract key information (slots) from their input.

...

To complete these goals you must go through all subpages in this section starting with Ontology.

...

Repository Overview

Note

Find the Intent and Slot Classifier files under /pca-agent-2025/social-interaction-cloud/sic_framework/services/nlu/utils

Code Block
breakoutModewide
The repository is structured as follows:
utils/
├── checkpoints/          # Directory for saving trained model checkpoints
├── data/                 # Directory containing data files (ontology, train/test datasets)
│   ├── ontology.json     # Ontology file containing intents, slots, and synonyms
│   ├── train.json        # Training dataset
│   ├── test.json         # Test dataset
│   └── synonyms.json     # Synonyms for slot normalization
├── data_processing.py    # Utilities for additional data preprocessing (if needed)
├── dataset.py            # Dataset preparation and preprocessing module
├── evaluation.py         # Model evaluation and metrics generation
├── run_train_test.py     # Main script to run training, evaluation, and inference
├── model.py              # Defines the BERT-based model architecture
├── predict.py            # Inference module for predicting intents and slots
├── requirements.txt      # Python dependencies for the project
├── train.py              # Training module for the intent-slot classifier
└── utils.py              # Helper functions for argument parsing, slot normalization, and synonym resolution

...

Argument

Type

Default

Description

--ontology_path

str

./data/ontology.json

Path to the ontology JSON file.

--train_data

str

./data/train.json

Path to the training dataset.

--test_data

str

./data/test.json

Path to the test dataset.

--model_save_path

str

checkpoints/model_checkpoint.pt

Path to save/load the trained model weights.

--train_model

bool

False

Train the model when this flag is set.

--evaluate

bool

False

Evaluate the model on the test dataset when this flag is set.

--num_epochs

int

2

Number of epochs for training.

--batch_size

int

16

Batch size for training.

--learning_rate

float

5e-5

Learning rate for the optimizer.

--max_length

int

16

Maximum sequence length for tokenization.

--seed

int

42

Random seed for reproducibility.

--inference_text

str

None

Text input for running inference.

--show_dist

bool

False

Show the intent and slot distribution in the dataset.

...

Use

...

Info

Due to how the imports are set up and because the intent and slot classifier is part of the social-interaction-cloud python package when you make changes you need to make sure to reinstall the social interaction cloud via pip install .

1. Viewing Dataset Distribution

...

Code Block
python run_train_test.py --show_dist

...

2. Training the Model

To train the model using the training dataset:

Code Block
python mainrun_train_test.py --train_model

...

3. Evaluating the Model

To evaluate a pre-trained model on the test dataset:

Code Block
python mainrun_train_test.py --evaluate

...

4. Running Inference

To predict the intent and slots for a given input text:

Code Block
python mainrun_train_test.py --inference_text "Find me a Japanese recipe for lunch."

5. Run with ASR

When complete with this section read https://socialrobotics.atlassian.net/wiki/spaces/PCA2/pages/2709488567/Run+your+Conversational+Agent#Run-your-Intent-and-Slot-Classifier-with-WHISPER to connect your intent and slot classifier with WHISPER